There is considerable evidence that treatment for alcohol and drug disorders can lead to significant improvements in substance use and psychosocial problem severity. Unfortunately, many patients relapse to prior levels of use relatively quickly after treatment has ended. Patients are therefore frequently referred to aftercare programs in an effort to maintain the improvements brought about by rehabilitation treatment. The efficacy of aftercare programs for cocaine patients has received very little empirical study, however. We propose to randomly assign cocaine patients entering the aftercare program at the VAMC into one of two conditions: (1) Standard aftercare drug counseling (two groups per week), and (2) Relapse prevention (one individual RP session plus one group session per week). Each program will last for 5 months and will be followed by 12-16 months of once per week group therapy. The relative efficacy and cost effectiveness of each condition will be evaluated at specific follow-up intervals up to 24 months post intake. Follow-ups will include assessments of substance use (self-report and urine toxicology), psychosocial problem severity, high risk behaviors for AIDS and utilization of health and social services. A number of potential patient-treatment matches, identified in previous research, will also be evaluated to determine which patients do best in each treatment condition. Finally, psychopathology, severity of dependence, social support, mood, self-help group beliefs, participation, self-efficacy, and coping behaviors will be assessed at various points in the follow-up to determine whether these factors predict treatment response and outcome status.